CN103829966A - Method and system for automatically determining positioning line in detection image - Google Patents
Method and system for automatically determining positioning line in detection image Download PDFInfo
- Publication number
- CN103829966A CN103829966A CN201210489837.8A CN201210489837A CN103829966A CN 103829966 A CN103829966 A CN 103829966A CN 201210489837 A CN201210489837 A CN 201210489837A CN 103829966 A CN103829966 A CN 103829966A
- Authority
- CN
- China
- Prior art keywords
- scanned
- threedimensional model
- detecting image
- position line
- parameter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Abstract
The invention relates to a method and system for automatically determining a positioning line in a detection image. The method comprises the obtaining step of obtaining a three-dimensional model which is in best fit with an object to be scanned and the determining step of projecting the positioning line in the best-fit three-dimensional model to the detection image of the object to be scanned, and therefore obtaining the best positioning line. The system comprises obtaining equipment used for obtaining the three-dimensional model which is in best fit with the object to be scanned and determining equipment used for projecting the positioning line in the best-fit three-dimensional model to the detection image of the object to be scanned, and therefore the best positioning line is obtained.
Description
Technical field
The application relates to medical science detection field, relates in particular to the method and system of automatically determining position line at the detecting image detecting for medical science.
Background technology
In medical science detection field, for example, in carrying out as the detection of computed tomography (CT) or nuclear magnetic resonance, NMR (MR), conventionally need to determine sweep limits.This can realize by obtaining the detecting image of object to be scanned and on detecting image, suitable position line being set.But existing position line setting up procedure is to realize by operator's manual operation substantially, whole process effort is consuming time again, and the position of position line and the precision of angle be difficult to ensure, thereby causes being difficult to obtain sweep limits accurately.
As an example of complete Cranial Computed Tomography scanning example, existing workflow is described below.
For complete Cranial Computed Tomography scanning, the workflow of current employing mainly depends on manual operation.Fig. 1 illustrates this workflow.In step S110, patient is placed on table top.In step S120, manually adjust table top to make laser rays pass patient's head, and manually adjust the attitude of patient's head so that patient's orbitomeatal line meets laser rays, or manually make CT rack inclining so that laser rays meets patient's orbitomeatal line.No matter be manually to adjust the attitude of patient's head or manually make CT rack inclining be consistent with laser rays with the orbitomeatal line that makes patient, these are all effort and manual operation consuming time, and are not easy to obtain orbitomeatal line aligning accurately.Therefore, may need the adjustment of repetition.This manual operation is commonly referred to attitude adjustment.
After having carried out attitude adjustment, by the detecting scanning of carrying out as shown in step S130.In step S130, the sweep limits of detecting scanning is estimated, and should at least be covered whole head scope.Obtaining after detecting image, in step S140, user need to manually adjust based on this detecting image position and the angle of orbitomeatal line.Particularly, by manual click or drag mouse and change original position, end position and the angle of one group of scanning, and adjust sweep length according to head image.This manual operation is commonly referred to position line adjustment.
If before scanning to detecting after attitude adjustment, patient's head moves, user need to reset the inclination angle of CT frame, or readjusts the attitude of patient's head.
In step S150, sweep limits is manually set.Shown in Fig. 3 is the schematic diagram of sweep limits.As shown in Figure 3, the length of sweep limits is the extremely position, the crown shown in left-hand broken line of orbitomeatal line from being shown in dotted line as right side, and width is the head width parallel with orbitomeatal line.
For example,, head is scanned when being all suitable at all key parameters (, the position of orbitomeatal line and angle and sweep limits), for example axial scan or helical scanning, as shown in step S160.
In addition, also there is following problem for the current typical workflow of carrying out complete Cranial Computed Tomography scanning:
Manually laser rays or the plane of scanning motion being aimed at orbitomeatal line is to require great effort and process consuming time;
If operator is unfamiliar with, manually laser rays or the plane of scanning motion are aimed at possible inaccurate with orbitomeatal line;
Sometimes the angle of orbitomeatal line tilts to eye very much, can cause unnecessary eyeball part to be scanned; And
Sometimes the angle of orbitomeatal line tilts or moves to cerebellum to cerebellum very much, can cause Brain not organized and not be scanned, and may need thus need to carry out multiple scanning or complementary scan according to doctor.
Recently, some methods based on image registration are studied.These methods based on image registration attempt automatically to detect and locate orbitomeatal line.For example, but image registration is not suitable for the unsharp detecting image of profile, the detecting image of head.For example the detecting image of CT detecting image comprises the three-dimensional information overlapping on two dimensional surface, and three-dimensional information amount is huge, may lack sharp-pointed and edge and may there is no significant characteristic point clearly.Therefore, till now, also do not have suitable method for registering images can be successfully used to automatically detect and locate the orbitomeatal line in CT detecting image.
Therefore, need a kind of method and system, in detecting image, automatically determine position line to realize.
Summary of the invention
For solving the above-mentioned problem, the application provides a kind of method and system of automatically determining the position line in detecting image.The three-dimensional information that the method and system utilization comprise in two dimension detecting image automatically detects and locates position line.In complete scan protocols, after detecting scanning, the present invention has been the process of full automatic treatment.
For example, in complete Cranial Computed Tomography scan protocols, the invention provides a kind of method and system of automatically determining orbitomeatal line, this has been the process of full automatic treatment.
It is a kind of for automatically determining the method for position line of detecting image that the application provides, and comprising: obtaining step, for obtaining and the threedimensional model of object to be scanned optimum matching; And determining step, for the position line of the threedimensional model of described optimum matching is projected in the detecting image of described object to be scanned, thereby obtain best position line.
According in the method for one or more embodiment of the present invention, described obtaining step comprises: calculation procedure, for calculating the matching error between the detecting image of the threedimensional model corresponding with described object to be scanned and the detecting image of described object to be scanned; And set-up procedure, for adjusting the described threedimensional model corresponding with described object to be scanned so that described matching error minimum, wherein, when described matching error hour, corresponding threedimensional model is the model of described optimum matching.
According in the method for one or more embodiment of the present invention, the described threedimensional model corresponding with described object to be scanned is to build in advance according to the meansigma methods that belongs to the image slice of many groups of scannings of the object in same ethnic group or region with described object to be scanned.
According in the method for one or more embodiment of the present invention, described set-up procedure comprises: setting steps, for arranging one group for adjusting the parameter of the described threedimensional model corresponding with described object to be scanned; And optimization step, for described one group of parameter being optimized so that described matching error minimum by optimized algorithm.
According in the method for one or more embodiment of the present invention, described one group of parameter comprises: mould shapes size parameter, for changing shape and/or the size of the described threedimensional model corresponding with described object to be scanned; Model attitude parameter, for changing the attitude of the described threedimensional model corresponding with described object to be scanned; The set of special shape control point, for realizing the special shape of the described threedimensional model corresponding with described object to be scanned; And/or translation parameters, for the described threedimensional model corresponding with described object to be scanned carried out to translation.
According in the method for one or more embodiment of the present invention, described optimized algorithm comprises Gauss-Newton algorithm, Levenberg-Marquardt algorithm or other applicable optimized algorithm.
According in the method for one or more embodiment of the present invention, also comprise according to the position line of described the best and sweep limits is set for follow-up scanning, for example axial scan or helical scanning.
According in the method for one or more embodiment of the present invention, also comprise and carry out as required translation and/or rotate the position line of described the best and sweep limits is set for follow-up scanning, for example axial scan or helical scanning according to translation and/or postrotational position line.
According in the method for one or more embodiment of the present invention, described object to be scanned has clear and definite bone shape structure, for example head or lumbar vertebra.
According in the method for one or more embodiment of the present invention, described position line comprises orbitomeatal line and/or basis cranii baseline.
According in the method for one or more embodiment of the present invention, described detecting image is the detecting image for computed tomography or nuclear magnetic resonance, NMR.
It is a kind of for automatically determining the system of position line of detecting image that the application also provides, and comprising: obtain equipment, for obtaining and the threedimensional model of object to be scanned optimum matching; And definite equipment, for the position line of the threedimensional model of described optimum matching is projected in the detecting image of described object to be scanned, thereby obtain best position line.
According in the system of one or more embodiment of the present invention, described in obtain equipment and comprise: accountant, for calculating the matching error between the detecting image of the threedimensional model corresponding with described object to be scanned and the detecting image of described object to be scanned; And adjusting device, for adjusting the described threedimensional model corresponding with described object to be scanned so that described matching error minimum, wherein, when described matching error hour, corresponding threedimensional model is the model of described optimum matching.
According in the system of one or more embodiment of the present invention, the described threedimensional model corresponding with described object to be scanned is to build in advance according to the meansigma methods that belongs to the image slice of many groups of scannings of the object in same ethnic group or region with described object to be scanned.
According in the system of one or more embodiment of the present invention, described adjusting device comprises: parts are set, for arranging one group for adjusting the parameter of the described threedimensional model corresponding with described object to be scanned; And optimization component, for described one group of parameter being optimized so that described matching error minimum by optimized algorithm.
According in the system of one or more embodiment of the present invention, described one group of parameter comprises: mould shapes size parameter, for changing shape and/or the size of the described threedimensional model corresponding with described object to be scanned; Model attitude parameter, for changing the attitude of the described threedimensional model corresponding with described object to be scanned; The set of special shape control point, for realizing the special shape of the described threedimensional model corresponding with described object to be scanned; And/or translation parameters, for the described threedimensional model corresponding with described object to be scanned carried out to translation.
According in the system of one or more embodiment of the present invention, described optimized algorithm comprises Gauss-Newton algorithm, Levenberg-Marquardt algorithm or other applicable optimized algorithm.
According in the system of one or more embodiment of the present invention, also comprise according to the position line of described the best the parts of sweep limits for follow-up scanning, for example axial scan or helical scanning are set.
According in the system of one or more embodiment of the present invention, also comprise and carry out as required translation and/or rotate the position line of described the best and according to translation and/or postrotational position line, the parts of sweep limits for follow-up scanning, for example axial scan or helical scanning are set.
According in the system of one or more embodiment of the present invention, described object to be scanned has clear and definite bone shape structure, for example head or lumbar vertebra.
According in the system of one or more embodiment of the present invention, described position line comprises orbitomeatal line and/or basis cranii baseline.
According in the system of one or more embodiment of the present invention, described detecting image is the detecting image for computed tomography or nuclear magnetic resonance, NMR.
The present invention provides again a kind of computed tomograph scanner system or NMR system, comprises the system of the position line for automatically determining detecting image.
The present invention at least provides following advantage:
The process of complete full automatic treatment can be avoided the manual operation consuming time of requiring great effort again, has simplified the workflow that medical science detects, thereby has improved the efficiency that medical science detects;
Improve the position of position line and the precision of angle, presented better performance thereby medical science is detected;
Farthest avoid unnecessary part to be scanned, thereby the dosage of X ray is reduced; And
For meeting market and clinical needs, can design different use patterns according to user operation habits.
Brief description of the drawings
By in conjunction with the following drawings, and with reference to the following detailed description to detailed description of the invention, can there is more thorough understanding to the present invention.
Shown in Fig. 1 is the schematic diagram for the work at present flow process of complete Cranial Computed Tomography scanning.
Shown in Fig. 2 is the schematic diagram of the improved workflow for complete Cranial Computed Tomography scanning according to an embodiment of the invention.
Shown in Fig. 3 is the schematic diagram of sweep limits.
Shown in Fig. 4 is the schematic diagram that extracts the process of three-dimensional skull model according to CT axial scan image sequence.
Shown in Fig. 5 is the schematic diagram that three-dimensional head model is carried out to the process of the simulation detecting projection of some angles according to an embodiment of the invention.
Shown in Fig. 6 is according to the schematic diagram of the process of the orbitomeatal line for automatically definite CT detecting image of an embodiment of invention.
Shown in Fig. 7 is the schematic diagram of the input and output of optimized algorithm according to an embodiment of the invention.
Shown in Fig. 8 is the schematic diagram that three kinds of designing according to user preference according to an embodiment of the invention use pattern.
Detailed description of the invention
By accompanying drawing, as an example instead of restriction embodiments of the invention as herein described are described.For illustrate succinct and clear for the purpose of, the element shown in figure is not necessarily drawn in proportion.For example, for the sake of clarity, the size of some elements may be with respect to other element through amplifying.In addition, in the situation that thinking fit, repeat reference numerals in accompanying drawing, to represent corresponding or similar element.In description, mentioning " embodiment " of the present invention or " embodiment " represents to comprise at least one embodiment of the present invention in conjunction with specific features, structure or characteristic described in this embodiment.Therefore, word " in one embodiment " differs and establishes a capital the same embodiment of expression in the appearance of each position of this description.
Below will be as an example of Cranial Computed Tomography scanning example come that the present invention is described in detail.It should be noted that the present invention is not limited to head.All be applicable to the present invention, for example lumbar vertebra as long as thering is clear and definite organizing of bone shape structure.In addition, it is also to be noted that, the present invention is not limited to CT scan, is also applicable to other situation of for example nuclear magnetic resonance, NMR.
Automatically the method that detects and locate the position line of for example orbitomeatal line in Cranial Computed Tomography scanning in the two dimension detecting image of head will be illustrated in below.It should be noted that the present invention is not limited to automatic detection and the location of orbitomeatal line, is also applicable to automatic detection and the location of other position line of for example basis cranii baseline.The method can be utilized the three-dimensional information comprising in two dimension detecting image, and in complete Cranial Computed Tomography scan protocols, the method is the process of the complete full automatic treatment after detecting scanning and before axial scan or helical scanning.
Shown in Fig. 1 is the schematic diagram for the work at present flow process of complete Cranial Computed Tomography scanning.As previously described, step S120, step S140 and step S150 are all manual operation processes, require great effort consuming time again, and the position of the orbitomeatal line obtaining and the precision of angle are difficult to ensure.
Shown in Fig. 2 is the schematic diagram of the improved workflow for complete Cranial Computed Tomography scanning according to an embodiment of the invention.Workflow shown in Fig. 1 and Fig. 2 is contrasted, can find out, workflow according to an embodiment of the invention is quite simple, because the manual operation shown in three step S120, S140 in original workflow as shown in Figure 1, S150 is replaced by the process of the complete full automatic treatment on CT control station completely, the process of this complete full automatic treatment is as shown in the step S270 in Fig. 2.In step S270, automatically determine position and angle and the sweep limits of orbitomeatal line.Particularly, the manual attitude adjustment in original workflow, manual positioning line adjustment and the adjustment of manual scanning scope replace with according to the automatic deterministic process of the position of the orbitomeatal line in workflow of the present invention and angle and sweep limits.Automatically determining according to the present invention behind the position and angle and sweep limits of orbitomeatal line, by the position of orbitomeatal line and angle and sweep limits Lookup protocol in head CT scan agreement for follow-up scanning, for example axial scan or helical scanning.
Step S210, S230, S260 in Fig. 2 are identical with step S110, S130, S160 in Fig. 1, repeat no more here.As shown in step S230, obtaining after real head detecting image, carrying out the process of the complete full automatic treatment as shown in step S270.In the process of this complete full automatic treatment, automatically determine position and angle and the sweep limits of orbitomeatal line, and automatically the position of determined orbitomeatal line and angle and sweep limits are set in Cranial Computed Tomography scan protocols for follow-up scanning to for example axial scan or helical scanning.
In the method for automatically definite orbitomeatal line according to the present invention, can use three-dimensional variable head model.Can build in advance three-dimensional variable head model.Once successfully build three-dimensional variable head model, can directly use it in clinical practice, only need on CT control station, move the software that is used for realizing the method according to this invention and automatically calculate optimum model parameter and mate best clinical case with the shape and the attitude that make model.
The axial scan image sequence that the complete head scanning that can carry out according to the case by previous obtains extracts three-dimensional skull model point, as shown in Figure 4.Threshold value T is rule of thumb set, thereby the pixel of each CT section is divided into two parts, i.e. bone parts and non-bone parts.In Fig. 4, shown in Fig. 4 a is CT axial scan image sequence, and shown in Fig. 4 b is the CT axial scan image sequence based on shown in Fig. 4 a and the three-dimensional skull model point cloud that forms.
Consider between the three-dimensional variable skull model of different ethnic groups and may have sizable shape difference, can plant or the ethnic group in each region builds a kind of model for everyone.Every kind of model builds by corresponding average case.General difference in every kind of model can compensate by the distortion of model.
If can not cover well special individual shapes by the general distortion of model, can in model, design special control point set.For example, can design two control point for cheekbone.Promote these two control point (three-dimensional point is around by carrying out weighting and also can be raised with the distance at control point) and can cover abnormal high-malar case.
After obtaining the threedimensional model of skull, can utilize simulation detecting projection algorithm to simulate detecting projection to the threedimensional model of this skull, thereby the detecting image that obtains simulation is for mating with real detecting image.This simulation detecting projection algorithm simulation CT detecting scanning.The detailed process of simulation detecting projection algorithm according to an embodiment of the invention is as follows.
In three-dimensional system of coordinate xyz, suppose that the central point of threedimensional model is positioned at coordinate (x
1, y
1, z
1), point source is positioned at coordinate (x
2, y
2, z
2).Point source irradiates vertically downward this threedimensional model in the direction of z axle, and like this, the fan beam projection line that point source produces is through this threedimensional model, thus the detecting image that generation is simulated in xoy plane.The fan beam projection line producing according to point source, through the different directions of threedimensional model, can produce the detecting image of the different angles of simulation, for example, and 0 degree of simulation or 90 degree detecting images, as shown in Figure 5.Conventionally, 0 degree detecting image of simulation obtains by front projection, and 90 degree detecting images of simulation obtain by side projection.In Fig. 5, shown in Fig. 5 a is 0 degree detecting projection, i.e. front projection of the simulation that obtains according to three-dimensional head model; Shown in Fig. 5 b is 90 degree detecting projection, i.e. side projections of the simulation that obtains according to three-dimensional head model; Shown in Fig. 5 c is the detecting projection of any attitude of the simulation that obtains according to three-dimensional head model.
It should be noted that and can carry out to three-dimensional variable head model the simulation detecting projection of various angles, to obtain the detecting image of simulation of various angles.Then, the detecting image of the simulation of obtained various angles can be mated with the real detecting image of the corresponding angle of clinical case.Various angles can be for example 0 degree, 90 degree or other is arbitrarily angled.
Successfully building after three-dimensional variable head model, can in three-dimensional variable head model, define orbitomeatal line accurately.Be projected in the detecting image of corresponding simulation at the orbitomeatal line defining in model shown in Fig. 5 b and Fig. 5 c.When by with the three-dimensional head model projection of clinical case optimum matching on two-dimensional ct detecting image time, the orbitomeatal line of the simulation in the three-dimensional head model of this optimum matching also can be mapped on two-dimensional ct detecting image, thereby obtain best orbitomeatal line.
In order automatically to determine the position line in detecting image, can be out of shape three-dimensional variable model, rotation and/or translation with clinical case optimum matching.Particularly, the detecting of three-dimensional variable modeling is projected in two dimensional surface to obtain the detecting image of simulation.The detecting image of simulation is mated with real detecting image.In the time obtaining with the shape of clinical case optimum matching and attitude parameter, the position line in the threedimensional model of optimum matching is projected on real detecting image.Then, calculate the parameter for follow up scan, comprise position and angle and the sweep limits of position line.
According to one embodiment of present invention, in order automatically to determine the orbitomeatal line in CT detecting image, can be out of shape three-dimensional variable head model, rotation and/or translation with clinical case optimum matching.Particularly, three-dimensional variable head model simulation detecting is projected in two dimensional surface to obtain the detecting image of simulation.The detecting image of simulation is mated with CT detecting image.In the time obtaining with the shape of clinical case optimum matching and attitude parameter, the orbitomeatal line in the three-dimensional head model of optimum matching is projected on CT detecting image.Then, calculate the parameter for CT follow up scan, comprise position and angle and the sweep limits of orbitomeatal line.
Below by according to an embodiment of the invention by three-dimensional variable head model and mating of clinical case, be optimized and automatically determine that the method for orbitomeatal line is described in detail.
For specific clinical case, change and/or make model generation translation with the special shape that shape, size, attitude and/or the control point of for example model are described by changing model parameter.
Shown in Fig. 6 is according to the schematic diagram of the method for the orbitomeatal line for automatically definite CT detecting image of an embodiment of invention.
As previously described, after successfully building three-dimensional variable head model, can make the shape of three-dimensional variable head model and/or 3 d pose change and/or make three-dimensional variable head model generation translation by a group model parameter.What in the time that the method shown in Fig. 6 starts, use is initial model parameter sets, and this initial model parameter sets is by vector
x inithe parameter sets representing, comprises following parameter:
Shape size parameter: Sx, Sy, Sz;
Special Shape Control Point set: Ss;
Model attitude parameter: Rx, Ry, Rz; And/or
Translation parameters: Tx, Ty.
In above-mentioned parameter, shape size parameter is for changing shape and/or the size of model to be optimized; Model attitude parameter is for changing the attitude of model to be optimized; Special Shape Control Point set is for realizing the special shape of model to be optimized; Translation parameters is for carrying out translation by model to be optimized.
Initial model parameter sets is rule of thumb set to represent average case.
In one or more embodiment according to the present invention, shape size parameter can be for representing model to be optimized to carry out the ratio of convergent-divergent.For example, work as Sx=0.7, Sy=0.7, when Sz=0.7, represents that by the size reduction of model to be optimized be original 70%.
In one or more embodiment according to the present invention, model attitude parameter can be for the angle that represents model to be optimized to be rotated.For example, when Rx=30 degree, Ry=40 degree, when Rz=50 spends, represents model to be optimized to turn clockwise in the direction of x axle 30 degree, 40 degree that turn clockwise in the direction of y axle, and 50 degree turn clockwise in the direction of z axle.
In one or more embodiment according to the present invention, special Shape Control Point set can comprise two control point for cheekbone.Promote these two control point (three-dimensional point is around by carrying out weighting and also can be raised with the distance at control point) and can cover abnormal high-malar case.
In one or more embodiment according to the present invention, translation parameters can be for the distance that represents model to be optimized to be carried out in xoy plane to translation.For example, work as Tx=5mm, when Ty=10mm, represent model to be optimized 5 millimeters of translations in the positive direction of x axle, 10 millimeters of translations in the positive direction of y axle.
Should be understood that the shape size parameter that model parameter is not limited to provide: Sx, Sy, Sz above; Special Shape Control Point set: Ss; Model attitude parameter: Rx, Ry, Rz; And translation parameters: Tx, Ty.For example, in one or more embodiment according to the present invention, translation parameters can be Tx, and Tz, for representing model to be optimized to carry out in xoz plane the distance of translation.In addition, special Shape Control Point set is also not limited to the control point for cheekbone above-mentioned.
As previously described, can carry out to threedimensional model the simulation detecting projection of various angles, thereby obtain the detecting image of the simulation of corresponding angle.Various angles can be 0 degree, 90 degree or other is arbitrarily angled.Then, the detecting image of simulation is mated with corresponding real detecting image, matching error is
e.According to one embodiment of present invention, minimize matching error by optimized algorithm in the mode of iteration
e, obtain
e min thereby, obtain final model parameter set
x opt.Final model parameter set
x optcorresponding mould shapes and attitude and actual clinical case optimum matching.
To the method shown in Fig. 6 be described in detail below.
In step S610, the initial model parameter sets of three-dimensional variable head model is rule of thumb set
x inito represent average case.Initial model parameter sets
x inicomprise Sx, Sy, Sz; Ss; Rx, Ry, Rz; Tx, Ty.90 degree detecting images 1010 of the simulation of three-dimensional variable head model in initial attitude 1000 and its correspondence are also shown in Fig. 6.90 degree detecting images 1010 of simulation are that the simulation detecting projection by three-dimensional variable head model 1000 being carried out to 90 degree produces.
In step S620, calculate 90 degree detecting images 1010 and real 90 matching degrees of spending between detecting images 1100 of simulation.
In step S630, utilize optimized algorithm to solve the head model parameter of mating with two-dimensional ct detecting image optimum.Shown in Fig. 7 is the schematic diagram of the input and output of optimized algorithm according to an embodiment of the invention.As shown in Figure 7, using head model parameter to be optimized, detect the detecting image of image and corresponding simulation as the input of optimized algorithm really, head model parameter to be optimized is optimized in the mode of iteration by optimized algorithm, finally obtains the head model parameter of mating with two-dimensional ct detecting image optimum.In one embodiment of the invention, 90 degree of head model parameter to be optimized, real 90 degree detecting images, simulation can be detected to the input of images as optimized algorithm, head model parameter to be optimized is optimized in the mode of iteration by optimized algorithm, thereby obtains the head model parameter of mating with real 90 degree detecting image optimums.In another embodiment of the present invention, 0 degree and 90 of head model parameter to be optimized, real 0 degree and 90 degree detecting images, simulation can be spent to the input of detecting images as optimized algorithm, head model parameter to be optimized is optimized in the mode of iteration by optimized algorithm, thereby obtains and the head model parameter of actual clinical case optimum matching.Be understandable that, the input of optimized algorithm is not limited to above-mentioned example.Can input the true detecting image of head model parameter to be optimized and various angles and simulation detecting image for obtaining and the head model parameter of actual clinical case optimum matching to optimized algorithm.
Can use for example Gauss-Newton algorithm, Levenberg-Marquardt algorithm or other applicable optimized algorithm, carry out Optimized model parameter sets with iterative manner.By minimizing matching error
e, obtain
e min thereby, obtain final model parameter set
x opt.Final model parameter set
x optcorresponding mould shapes and attitude and actual clinical case optimum matching.
When find final model parameter set in step S630
x optafterwards, in step S640, by with final model parameter set
x optthe orbitomeatal line of the simulation in the variable header model that the real detecting image optimum of corresponding and actual clinical case mates projects in CT detecting image, thereby obtains best orbitomeatal line.
The orbitomeatal line of final model is projected on CT detecting image, obtain position and the angle of best orbitomeatal line.After having determined the position and angle of best orbitomeatal line, can determine the width from best orbitomeatal line to the crown, i.e. sweep limits.Like this, just obtained the sweep parameter for full header scanning subsequently, i.e. the position of orbitomeatal line and angle and sweep limits.Then by the position of the orbitomeatal line obtaining and angle and sweep limits Lookup protocol in head CT scan agreement for follow-up scanning, for example axial scan or helical scanning.
In Fig. 6,90 degree detecting images 1030 and the real 90 degree detecting images 1110 of the simulation of the three-dimensional variable head model of the three-dimensional variable head model 1020 that mates with real detecting image optimum, optimum matching are also shown.In the three-dimensional variable head model 1020 of optimum matching, show the orbitomeatal line 1025 of simulation.In real 90 degree detecting images 1110, show obtained best orbitomeatal line 1115 and by the sweep limits shown in rectangle frame 1118.
In order to meet market and clinical needs, can design different use patterns according to user operation habits.Shown in Fig. 8 is the schematic diagram that three kinds of designing according to user preference according to an embodiment of the invention use pattern, comprises Fig. 8 a, Fig. 8 b and Fig. 8 c.
Shown in Fig. 8 a is normal orbitomeatal line pattern, and the angle of wherein cutting into slices and dissection constructor close the existing knowledge structure of doctor, but part eyeball can be scanned.
Shown in Fig. 8 b is that eye is avoided pattern 1, wherein orbitomeatal line is rotated above eye to suitable angle, for example 5 degree-10 degree, and section angle is slightly different from the existing knowledge structure of doctor with anatomical structure.
Shown in Fig. 8 c is that eye is avoided pattern 2, and wherein by the upwards suitable distance of translation of orbitomeatal line, for example 5 millimeters-10 millimeters, section angle and dissection constructor close the existing knowledge structure of doctor, but part cerebellum is not scanned.
In some hospitals, use eye to avoid pattern 1.In some cases, to avoid pattern 2 can be preferred to eye.
In actual clinical practice, alternative use pattern is not limited to as above three kinds.Different use patterns can also be set according to actual needs to be selected for user.
The process of the complete full automatic treatment of the orbitomeatal line for automatically definite Cranial Computed Tomography detecting image according to an embodiment of the invention can at least realize following advantage:
Avoid the manual operation consuming time of requiring great effort again, simplified the workflow of Cranial Computed Tomography scanning, thereby improved the efficiency of Cranial Computed Tomography scanning;
Improve the position of orbitomeatal line and the precision of angle, thereby made Cranial Computed Tomography scanning present better performance;
Farthest avoid unnecessary eyeball part to be scanned, thereby the dosage of X ray is reduced; And
For meeting market and clinical needs, can design different use patterns according to user operation habits.
By specific embodiment, the present invention is described in detail above, but the present invention is not limited to above-described embodiment.Without departing from the scope of the invention, can carry out various modifications and changes to the present invention.Scope of the present invention is limited by appended claims.
Claims (23)
1. for automatically determining a method for the position line of detecting image, comprising:
Obtaining step, for obtaining and the threedimensional model of object to be scanned optimum matching; And
Determining step, for the position line of the threedimensional model of described optimum matching is projected in the detecting image of described object to be scanned, thereby obtains best position line.
2. the method for claim 1, wherein described obtaining step comprises:
Calculation procedure, for calculating the matching error between the detecting image of the threedimensional model corresponding with described object to be scanned and the detecting image of described object to be scanned; And
Set-up procedure, for adjusting the described threedimensional model corresponding with described object to be scanned so that described matching error minimum,
Wherein, when described matching error hour, corresponding threedimensional model is the model of described optimum matching.
3. method as claimed in claim 2, wherein, the described threedimensional model corresponding with described object to be scanned is that basis builds in advance with the meansigma methods that described object to be scanned belongs to the image slice of many groups of scannings of the object in same ethnic group or region.
4. method as claimed in claim 2 or claim 3, wherein, described set-up procedure comprises:
Setting steps, for arranging one group for adjusting the parameter of the described threedimensional model corresponding with described object to be scanned; And
Optimization step, for being optimized so that described matching error minimum described one group of parameter by optimized algorithm.
5. method as claimed in claim 4, wherein, described one group of parameter comprises:
Mould shapes size parameter, for changing shape and/or the size of the described threedimensional model corresponding with described object to be scanned;
Model attitude parameter, for changing the attitude of the described threedimensional model corresponding with described object to be scanned;
The set of special shape control point, for realizing the special shape of the described threedimensional model corresponding with described object to be scanned; And/or
Translation parameters, for carrying out translation by the described threedimensional model corresponding with described object to be scanned.
6. the method as described in claim 4 or 5, wherein, described optimized algorithm comprises Gauss-Newton algorithm, Levenberg-Marquardt algorithm or other applicable optimized algorithm.
7. the method as described in any one in claim 1-6, wherein, also comprises according to the position line of described the best and sweep limits is set for follow-up scanning, for example axial scan or helical scanning.
8. the method as described in any one in claim 1-6, wherein, also comprise and carry out as required translation and/or rotate the position line of described the best and sweep limits is set for follow-up scanning, for example axial scan or helical scanning according to translation and/or postrotational position line.
9. the method as described in any one in claim 1-8, wherein, described object to be scanned has clear and definite bone shape structure, for example head or lumbar vertebra.
10. method as claimed in any one of claims 1-9 wherein, wherein, described position line comprises orbitomeatal line and/or basis cranii baseline.
11. methods as described in any one in claim 1-10, wherein, described detecting image is the detecting image for computed tomography or nuclear magnetic resonance, NMR.
12. 1 kinds of systems for the automatic position line of determining detecting image, comprising:
Obtain equipment, for obtaining and the threedimensional model of object to be scanned optimum matching; And
Determine equipment, for the position line of the threedimensional model of described optimum matching is projected in the detecting image of described object to be scanned, thereby obtain best position line.
13. systems as claimed in claim 12, wherein, described in obtain equipment and comprise:
Accountant, for calculating the matching error between the detecting image of the threedimensional model corresponding with described object to be scanned and the detecting image of described object to be scanned; And
Adjusting device, for adjusting the described threedimensional model corresponding with described object to be scanned so that described matching error minimum,
Wherein, when described matching error hour, corresponding threedimensional model is the model of described optimum matching.
14. systems as claimed in claim 13, wherein, the described threedimensional model corresponding with described object to be scanned is that basis builds in advance with the meansigma methods that described object to be scanned belongs to the image slice of many groups of scannings of the object in same ethnic group or region.
15. systems as described in claim 13 or 14, wherein, described adjusting device comprises:
Parts are set, for arranging one group for adjusting the parameter of the described threedimensional model corresponding with described object to be scanned; And
Optimization component, for being optimized so that described matching error minimum described one group of parameter by optimized algorithm.
16. systems as claimed in claim 15, wherein, described one group of parameter comprises:
Mould shapes size parameter, for changing shape and/or the size of the described threedimensional model corresponding with described object to be scanned;
Model attitude parameter, for changing the attitude of the described threedimensional model corresponding with described object to be scanned;
The set of special shape control point, for realizing the special shape of the described threedimensional model corresponding with described object to be scanned; And/or
Translation parameters, for carrying out translation by the described threedimensional model corresponding with described object to be scanned.
17. systems as described in claim 15 or 16, wherein, described optimized algorithm comprises Gauss-Newton algorithm, Levenberg-Marquardt algorithm or other applicable optimized algorithm.
18. systems as described in any one in claim 12-17, wherein, also comprise according to the position line of described the best the parts of sweep limits for follow-up scanning, for example axial scan or helical scanning are set.
19. systems as described in any one in claim 12-17, wherein, also comprise and carry out as required translation and/or rotate the position line of described the best and according to translation and/or postrotational position line, the parts of sweep limits for follow-up scanning, for example axial scan or helical scanning are set.
20. systems as described in any one in claim 12-19, wherein, described object to be scanned has clear and definite bone shape structure, for example head or lumbar vertebra.
21. systems as described in any one in claim 12-20, wherein, described position line comprises orbitomeatal line and/or basis cranii baseline.
22. systems as described in any one in claim 12-21, wherein, described detecting image is the detecting image for computed tomography or nuclear magnetic resonance, NMR.
23. 1 kinds of computed tomograph scanner systems or NMR system, comprise as described in any one in claim 12-22 for automatically determining the system of position line of detecting image.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210489837.8A CN103829966B (en) | 2012-11-27 | 2012-11-27 | For automatically determining the method and system of the position line in detecting image |
| US14/069,524 US9299148B2 (en) | 2012-11-27 | 2013-11-01 | Method and system for automatically determining a localizer in a scout image |
| JP2013242459A JP6313024B2 (en) | 2012-11-27 | 2013-11-25 | Method and system for automatically determining a localizer within a scout image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210489837.8A CN103829966B (en) | 2012-11-27 | 2012-11-27 | For automatically determining the method and system of the position line in detecting image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103829966A true CN103829966A (en) | 2014-06-04 |
| CN103829966B CN103829966B (en) | 2018-12-07 |
Family
ID=50773350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210489837.8A Active CN103829966B (en) | 2012-11-27 | 2012-11-27 | For automatically determining the method and system of the position line in detecting image |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US9299148B2 (en) |
| JP (1) | JP6313024B2 (en) |
| CN (1) | CN103829966B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127550A (en) * | 2019-12-26 | 2020-05-08 | 上海联影医疗科技有限公司 | Image slice positioning method, device and system based on bed board and storage medium |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102013220665A1 (en) * | 2013-10-14 | 2015-04-16 | Siemens Aktiengesellschaft | Determining a value of a recording parameter by means of an anatomical landmark |
| CN104905766B (en) * | 2014-03-11 | 2019-09-24 | Ge医疗系统环球技术有限公司 | Medical scanning system and method based on positioning as determining sweep parameter |
| WO2015168781A1 (en) * | 2014-05-06 | 2015-11-12 | Conceptualiz Inc. | System and method for interactive 3d surgical planning and modelling of surgical implants |
| WO2016006633A1 (en) * | 2014-07-10 | 2016-01-14 | 国立大学法人 鹿児島大学 | Diagnosis device, diagnosis system, simulation method, and program |
| JP7009345B2 (en) * | 2018-11-01 | 2022-01-25 | ユニ・チャーム株式会社 | Absorbent article |
| KR20200075623A (en) * | 2018-12-18 | 2020-06-26 | (주)제노레이 | Dental Treatment Planning Apparatus Using Matching Of 2D Medical Image And 3D Medical Image And Method Thereof |
| JP7222947B2 (en) * | 2020-04-01 | 2023-02-15 | 株式会社モリタ製作所 | X-ray imaging device and X-ray imaging method |
| DE102020213690B4 (en) | 2020-10-30 | 2022-01-20 | Siemens Healthcare Gmbh | Method for adjusting a shape of a headstock, method for providing imaging data and system with a headstock |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1736333A (en) * | 2004-07-21 | 2006-02-22 | Ge医疗系统环球技术有限公司 | Computed tomography dose indexing phantom selection for dose reporting |
| US20090129650A1 (en) * | 2007-11-19 | 2009-05-21 | Carestream Health, Inc. | System for presenting projection image information |
| US20090316966A1 (en) * | 2008-05-16 | 2009-12-24 | Geodigm Corporation | Method and apparatus for combining 3D dental scans with other 3D data sets |
| CN101689298A (en) * | 2006-12-22 | 2010-03-31 | 皇家飞利浦电子股份有限公司 | Imaging system and imaging method for imaging an object |
| CN101686825A (en) * | 2007-06-21 | 2010-03-31 | 皇家飞利浦电子股份有限公司 | Use the dynamic model adjustment to be used for the acquisition protocols of dynamic medical imaging |
| US20110245656A1 (en) * | 2010-03-31 | 2011-10-06 | Roland Bammer | Arterial spin labeling MRI-based correction factor for improving quantitative accuracy of blood flow and volume images |
| CN102365655A (en) * | 2009-06-10 | 2012-02-29 | 三菱电机株式会社 | Image matching device and device for positioning patient |
| US20120148131A1 (en) * | 2010-12-08 | 2012-06-14 | Radimetrics, Inc. | Generating a suitable model for estimating patient radiation dose resulting from medical imaging scans |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007058918A2 (en) * | 2005-11-11 | 2007-05-24 | Hologic Inc. | Estimating risk of future bone fracture utilizing three-dimensional bone density model |
| US8218849B2 (en) | 2008-09-04 | 2012-07-10 | Siemens Corporation | Method and system for automatic landmark detection using discriminative joint context |
| JP5121889B2 (en) * | 2010-07-22 | 2013-01-16 | ジーイー・メディカル・システムズ・グローバル・テクノロジー・カンパニー・エルエルシー | X-ray CT system |
| DE102010040634A1 (en) * | 2010-09-13 | 2012-03-15 | Siemens Aktiengesellschaft | Method for 2D / 3D registration |
-
2012
- 2012-11-27 CN CN201210489837.8A patent/CN103829966B/en active Active
-
2013
- 2013-11-01 US US14/069,524 patent/US9299148B2/en active Active
- 2013-11-25 JP JP2013242459A patent/JP6313024B2/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1736333A (en) * | 2004-07-21 | 2006-02-22 | Ge医疗系统环球技术有限公司 | Computed tomography dose indexing phantom selection for dose reporting |
| CN101689298A (en) * | 2006-12-22 | 2010-03-31 | 皇家飞利浦电子股份有限公司 | Imaging system and imaging method for imaging an object |
| CN101686825A (en) * | 2007-06-21 | 2010-03-31 | 皇家飞利浦电子股份有限公司 | Use the dynamic model adjustment to be used for the acquisition protocols of dynamic medical imaging |
| US20090129650A1 (en) * | 2007-11-19 | 2009-05-21 | Carestream Health, Inc. | System for presenting projection image information |
| US20090316966A1 (en) * | 2008-05-16 | 2009-12-24 | Geodigm Corporation | Method and apparatus for combining 3D dental scans with other 3D data sets |
| CN102365655A (en) * | 2009-06-10 | 2012-02-29 | 三菱电机株式会社 | Image matching device and device for positioning patient |
| US20110245656A1 (en) * | 2010-03-31 | 2011-10-06 | Roland Bammer | Arterial spin labeling MRI-based correction factor for improving quantitative accuracy of blood flow and volume images |
| US20120148131A1 (en) * | 2010-12-08 | 2012-06-14 | Radimetrics, Inc. | Generating a suitable model for estimating patient radiation dose resulting from medical imaging scans |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127550A (en) * | 2019-12-26 | 2020-05-08 | 上海联影医疗科技有限公司 | Image slice positioning method, device and system based on bed board and storage medium |
| CN111127550B (en) * | 2019-12-26 | 2023-08-29 | 上海联影医疗科技股份有限公司 | Image slice positioning method, device and system based on bed board and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| US20140147026A1 (en) | 2014-05-29 |
| JP6313024B2 (en) | 2018-04-18 |
| CN103829966B (en) | 2018-12-07 |
| US9299148B2 (en) | 2016-03-29 |
| JP2014104361A (en) | 2014-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103829966A (en) | Method and system for automatically determining positioning line in detection image | |
| Knörlein et al. | Validation of XMALab software for marker-based XROMM | |
| US10192325B2 (en) | Method for calibrating an imaging device and an imaging device | |
| US20210225003A1 (en) | Image processing method and apparatus, server, and storage medium | |
| CN107909622B (en) | Model generation method, medical imaging scanning planning method and medical imaging system | |
| CN110956633B (en) | Rapid CT scanning method and system based on virtual stereotactic image | |
| CN108053437B (en) | Three-dimensional model obtaining method and device based on posture | |
| EP3493516A1 (en) | Method and device for adjusting scanning state | |
| WO2023273093A1 (en) | Human body three-dimensional model acquisition method and apparatus, intelligent terminal, and storage medium | |
| US10169891B2 (en) | Producing three-dimensional representation based on images of a person | |
| CN110021053A (en) | A kind of image position method, device, storage medium and equipment based on coordinate conversion | |
| CN102831601A (en) | Three-dimensional matching method based on union similarity measure and self-adaptive support weighting | |
| KR102522873B1 (en) | Method for providing medical procedure result comparing service using 3-diementional face scan data | |
| CN111353506A (en) | Adaptive gaze estimation method and apparatus | |
| WO2009118020A1 (en) | 3d medical image segmentation | |
| US10438345B2 (en) | Improving symmetry in brain scans | |
| US11819427B2 (en) | Systems and methods for orthosis design | |
| CN110264527A (en) | Real-time binocular stereo vision output method based on ZYNQ | |
| JP2013220319A (en) | Image processing apparatus, method, and program | |
| AU2020217368A1 (en) | Technologies for determining the accuracy of three-dimensional models for use in an orthopaedic surgical procedure | |
| WO2015186439A1 (en) | Image processing device and three-dimensional display method | |
| CN109308472B (en) | Three-dimensional sight estimation method based on iris projection matching function | |
| JP2013175006A (en) | Device, method, and program for generating mesh | |
| CN104182963A (en) | Image processing method and image processing device | |
| US20140056499A1 (en) | Apparatus and method for generating image using correction model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |